Earthquake propagation
When a major earthquake breaks somewhere in New Zealand, the shock waves race outward across the country at almost four kilometres a second. From the epicentre to Wellington in under a minute. Right across both islands in three. This page replays that race for any of New Zealand’s biggest recent quakes — and for the latest significant one — as an animated map. Pick an event below and watch the waves pass through.
P-wave · cyan ring
The faster wave to arrive — felt as a deep boom or sharp crack, sometimes a building rumble. Animals notice it before humans do.
S-wave · orange ring
The slower, larger wave that arrives a few seconds later. This is the actual shaking — what breaks chimneys and cracks pavements.
What you’ve just watched
Every quake here actually happened. Position, depth, magnitude and origin time come straight from GeoNet, the public seismic monitoring network run by GNS Science. The rings are physics: P-waves travel through the Earth’s crust at roughly six kilometres per second, S-waves at about three and a half. The S-wave is the slower, larger-amplitude one — the shake that breaks chimneys and cracks pavements. As the wavefront sweeps across the country, towns in its path begin to feel the ground move; the map lights them up in the order they would have been shaken.
Real wave propagation is more complicated than constant-speed circles — geology refracts the waves, basin sediments amplify them, depth changes the pattern. But the timing is approximately right, and watching the country light up sequentially is the clearest way most people will ever see why a quake in Hanmer Springs shakes Wellingtonians out of bed before it reaches Christchurch’s southern suburbs.
After the main wave has cleared, the map continues into the aftershock sequence — every quake above the catalogue threshold that happened in the hours that followed, time-compressed so a full day passes in seconds. For the big ones — Kaikōura, Christchurch — that period is itself extraordinary to watch.
How earthquake waves travel
An earthquake is not a single instantaneous event but a tearing release of stored stress along a fault plane. The tear itself takes seconds to minutes — the M7.8 Kaikōura quake of 2016 unzipped along about a dozen faults over roughly seventy-five seconds. As the rock slips, it sends out two main kinds of seismic wave. The faster primary or P-wave is a longitudinal pressure wave, pushing and pulling the rock in the direction of travel; it moves through continental crust at about six kilometres per second and is what early-warning systems detect first. The slower secondary or S-wave is a transverse shear wave, oscillating perpendicular to its direction of travel; it moves at roughly three and a half kilometres per second. The S-wave carries most of the energy people perceive as shaking, and for any given location it arrives a few seconds to a few tens of seconds after the P-wave depending on distance from the epicentre.
Both wave types refract through the layered Earth, slowing in some rock types and speeding up in others. In the soft sedimentary basins underneath cities like Christchurch, Lower Hutt and parts of Wellington, the S-wave amplitude is amplified — sometimes by a factor of two or three — which is why some neighbourhoods experience much heavier shaking than the bedrock-founded ones nearby. The “ring” model on this map is an honest first approximation, but it is not the full picture; it is the kind of explanation a high-school physics teacher would give before getting into the geology.
What the P-wave actually feels like
The P-wave — the faster, smaller wave drawn in cyan on the map — isn’t dramatic ground motion. It’s something else: a deep low boom, a sharp crack, sometimes a building rumble that lasts a second or two. People near a sizeable quake will often hear the P-wave arrive before they feel anything. Doors and windows rattle briefly with no obvious cause, mistakable for a truck hitting the house, a sonic boom, or a distant explosion. Animals react before humans do — there are well-documented reports of dogs barking and birds taking off in the seconds before a quake is felt by people, and that’s the P-wave they’re picking up on. The higher-frequency edge of the P-wave’s energy spectrum overlaps with the audible range, so as it shakes windows, walls and the air itself, it produces sound.
In Christchurch in February 2011 a lot of residents reported hearing a sharp crack just before the shake hit. In Kaikōura in 2016 the same pattern — those close to the rupture zone heard a low building rumble for a second or two before everything started moving. Then the orange wave arrives, the ground starts rolling, and the actual shaking begins.
The interval between the P-wave and the S-wave arriving at a particular location is the basis of every modern earthquake early-warning system. Japan’s J-Alert and California’s ShakeAlert detect the P-wave at sensors near the epicentre, calculate from its first second of motion that something significant is coming, and push alerts to mobile phones, trains, and elevators before the destructive S-wave arrives. Tokyo can sometimes get thirty to sixty seconds of warning. New Zealand doesn’t yet have a national early-warning system at the J-Alert level; GeoNet has been trialling components of one but it is not yet operational at scale.
What the data says about NZ’s biggest quakes
The two events that defined New Zealand’s recent seismic memory are the 4 September 2010 Darfield earthquake in Canterbury and its 22 February 2011 Christchurch aftershock. The Darfield event was magnitude 7.1 and broke the Greendale Fault, a previously-unknown structure beneath the Canterbury Plains. It struck at 4:35am and although shaking was extreme, no one was killed — almost certainly because most people were in bed. The much-shallower February 2011 event was only magnitude 6.2, but the rupture happened directly beneath the south-east of central Christchurch with vertical accelerations approaching twice the force of gravity in places. 185 people died. The 14 November 2016 Kaikōura earthquake at magnitude 7.8 was even larger but offshore and rural; two people died, but the rupture itself — cascading across at least a dozen separate faults — is now used worldwide as a reference example of complex multi-fault behaviour.
Smaller events shake the country constantly. GeoNet records hundreds of quakes a week across New Zealand, the vast majority too small to be felt. The Newswire live earthquake map shows recent activity in close to real time; this page is its companion piece, focused on what happens in the seconds and hours after the very largest events.
Why aftershocks matter
The big rupture is rarely the end of the story. When a fault releases stress, the surrounding crust readjusts, often producing hundreds or thousands of smaller events in the days, weeks and months that follow. Most are far smaller than the main shock; a small fraction are large enough to do further damage. The Christchurch sequence produced over 13,000 recorded aftershocks, including more than 2,700 of magnitude 3 or above, in the year after Darfield. Several of them — including the February 2011 event itself — were significantly more destructive than the original quake.
The aftershock sequence on this map is filtered to the largest events for visual clarity (above magnitude 3.5 or 4 depending on the parent quake), and accelerated so a day passes in around fifteen seconds. Every flash is a real earthquake recorded by GeoNet’s seismograph network. Watch for the geographic pattern: aftershocks cluster along the rupture, gradually thin out, and occasionally migrate as stress redistributes onto adjacent faults.